Symptoms for which patients seek treatment can be related to three factors: the number of ways the patient pertitions his or her appraisal of issues in current interpersonal relations, the evaluative content of that appraisal, and the particular behavioral options that then seem to be available. Productive short-term psychodynamic treatment leads to symptom relief as well as change in these factors as measured by the Interpersonal Descrimination Task, the semantic differential rating of people important to the patient, and the Ways of Coping Checklist. Changes in these measures, we hypothesize, is dependent on working on key psychodynamic constructs: the resolution of the core conflictual relationship theme (CCRT) through its identification as a repetitive maladaptive pattern in relationship with others and understanding its operation in relation to the therapist and its effect on the therapeutic alliance. How resolution of the CCRT is carried out between patient and therapist will be studied using a newly developed content analysis measure: the Seattle Psychotherapy Language Schema (SPLASH). Six therapists will treat six patients for 15 sessions and six patients for 30 sessions for a total of 72 patients. Transcripts of psychotherapy sessions will be analyzed with the SPLASH. We will follow change in scores on the above-described measures, relating them to the amount and type of verbal interchange that occurs between patient and therapist as they work on the CCRTs.